14806
Blog

What Scaled, System-Wide AI Looks Like at University Hospitals

aiOS, System Integration, Workflow
Table of Contents

At University Hospitals (UH), innovation is not pursued for its own sake. It is applied in service of safety, quality and community health across Northeast Ohio. 

That approach is already producing measurable value for patients. Across 13 hospitals and more than 50 ambulatory centers, Aidoc is embedded into clinical workflow with a platform powered by 17 FDA-cleared AI algorithms. Across all institutions, 13 Aidoc algorithms are live, analyzing 1.3 million scans annually. 295 physicians actively use the platform, and more than 12,687 patients benefit each year from faster diagnosis and prioritization, alongside 2,700 hours of efficiency gains.

That track record shows how a community-rooted health system can adopt AI in a way that is integrated, measurable and sustainable. For Dr. Leonardo Kayat, that proven foundation is what makes the next phase of radiology possible: broader, smarter AI support for abdominal imaging.

Why abdominal imaging matters now

In a recent conversation, Dr. Kayat described abdominal imaging as one of the most challenging frontiers for AI. Unlike narrower use cases, the abdomen brings more organs, more disease states and more complexity to train and validate. While earlier AI development naturally focused on highly acute areas such as stroke, pulmonary embolism and vascular disease, abdominal imaging remained relatively underserved.

That is why this moment feels different. Dr. Kayat noted that foundation models are changing what is possible by making it feasible to cover many more abdominal conditions. For radiologists managing crowded worklists, that is not just a technical advance. It is a meaningful workflow improvement.

The real need is prioritization

At UH, radiologists balance a demanding environment: emergency department studies, urgent inpatient imaging and outpatient exams can all compete for attention. And when everything feels urgent, how do they decide which case to read first?

That is where Dr. Kayat sees the greatest value in abdominal AI today: triage, prioritization and notification. When STAT and ED exams must be turned around within an hour, outpatient backlogs inevitably grow — but if severe findings such as bowel ischemia are surfaced sooner, radiologists can act first on the cases that most urgently need attention. For subspecialists, that means better prioritization; for general radiologists, high-volume readers and trainees, it also provides an added layer of support.

Why UH stands out

UH has successfully operationalized AI at scale. Deep IT integration, strong clinical leadership and disciplined execution have helped embed Aidoc into real workflows rather than leaving it as siloed technology.

That matters because the future of radiology AI will  be shaped by comprehensive, system-wide integrations that support the realities of care delivery, physician workflow and care coordination. UH has already shown that the model can work.

Foundation models unlocking new capability

This is where the conversation moves beyond today’s algorithms. Aidoc’s foundation model work reflects a shift away from building one algorithm for one disease at a time. Instead, foundation models are designed to recognize patterns across many conditions simultaneously. 

For abdominal imaging, the need is especially clear to both clinicians and health system leaders. Clinically, the abdomen represents exactly the kind of complexity that makes traditional algorithm‑by‑algorithm development extremely challenging: it contains many more organs, compartments, and tissue types than the head or chest, and pathology covers a high number of more infrequent conditions. Taken together, these abdominal conditions account for high patient volume, but that volume is fragmented across numerous findings — making single‑purpose algorithms hard to justify and slow to develop.

From an operational and executive perspective, it wouldn’t make sense to build and maintain 14 separate point solutions just to cover the abdomen. That would take years to achieve meaningful coverage and leave IT and clinical teams managing 14 different integrations, workflows and devices. Foundation model technology changes that. It enables comprehensive coverage of abdominal CT through a single workflow, giving radiologists a broader safety net while offering leaders a more scalable, cost‑effective and easier‑to‑govern approach to AI deployment.

What radiologists want next

Dr. Kayat’s perspective also points to where AI should go next. The future is not only about flagging critical diseases and prioritizing those cases. It’s also about reducing the repetitive work that adds burden without adding value.

That vision is ultimately what makes this partnership meaningful. At UH, Aidoc is helping demonstrate that clinical AI gives clinicians better tools to prioritize, coordinate and practice at the top of their license.

Explore the Latest AI Insights, Trends and Research

Follow a manual added link

Webinars

Aidoc experts, customers and industry leaders share the latest in AI benefits and adoption.

Follow a manual added link

eBooks

Explore how clinical AI can transform your health system with insights rooted in real-world experiences.

Follow a manual added link

AI Strategy

Learn how to go beyond the algorithm to develop a scalable AI strategy and implementation plan.

Follow a manual added link

Solutions

Explore how Aidoc can help increase hospital efficiency, improve outcomes and demonstrate ROI.

© Copyright 2026 Aidoc